© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 13463
    Prev
    Next
    INDEX
    Explanations

    numerical data and reference points

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.47
    3.2%
    111
    +0.45
    3.1%
    478
    +0.09
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.47
    0.07
    156
    +0.45
    0.07
    71
    +0.09
    0.06
    Negative Logits
    thood
    -1.43
    inals
    -1.32
    erto
    -1.26
    ables
    -1.22
    ()</
    -1.19
     Territory
    -1.17
    ()`
    -1.16
    akes
    -1.16
    ennes
    -1.15
    ders
    -1.13
    POSITIVE LOGITS
    ·
    1.88
    ¶
    1.78
    į
    1.69
    Ĩ
    1.69
    ¡
    1.63
    ĨĴ
    1.61
    ³
    1.60
    ĻĤ
    1.52
    Īĺ
    1.51
    ÃĥÃĤÃĥÃĤ
    1.48
    Activations Density 0.055%

    No Known Activations